So for example have you noticed a big difference in support request volume on software being used by 1000 users compared to just 50? Or is it more related to how solid the software actually is? It's almost like calculating how valuable it is for us to take the extra time to make the software properly, which would be a nice thing to sell management on right guys?
views:
22answers:
1It depends a bit on the software. If it is designed to fill a very specific niche, then it can be well designed and keep most users happy. If it is highly configurable and could easily be extended in a million directions, there is more scope for user feedback. And of course bulletproof software gets far fewer reports than flakey software.
There are usually a limited number of core requests that you get from users (though the limit can be large for a complex program). After you hit a certain number of users, you tend to start getting duplicate requests rather than new issues, so the request rate tends to flatten out or stabilise as the user base increases.
The other effect is that early in the life cycle you tend to get more feedback, and as the product matures, the feedback rate tends to decrease as you nail the bugs and get the "missing" features in.
My direct experience only goes up to the hundreds of thousands of copies, though - different rules may come into play as you move into millions+.